Since when? Seems like he's always pulling made-up rumors out of his ass, and people just forget that he's wrong about everything until, as you point out, he gets actual information and changes his story at the last minute.
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
New Mac Studio Likely to Be Introduced at WWDC
- Thread starter MacRumors
- Start date
- Sort by reaction score
Seriously. Can we finally just all admit that he doesn't actually know anything? Why can't people remember this? He only gets things right when it's commonly known info.
He didn't know about the Mac Studio, and he doesn't know anything about the Mac Pro either. All his rumors are garbage.
I wondered if this would happen when it was rumored that the even higher end Mac Pro chip was cancelled for the M2. It just doesn't make sense to get a Mac Pro that has the same chip as the Ultra.
Hopefully we will see a Mac Pro with an M3 Extreme next year. I bet the die size for that thing on 5nm was just too crazy. 3nm should help if that's the case and the M3 is supposed to be on 3nm.
I'm thinking I'll probably upgrade my setup to a 14" M3 Max next year, especially if it has OLED. Should be quite the unit.
Hopefully we will see a Mac Pro with an M3 Extreme next year. I bet the die size for that thing on 5nm was just too crazy. 3nm should help if that's the case and the M3 is supposed to be on 3nm.
I'm thinking I'll probably upgrade my setup to a 14" M3 Max next year, especially if it has OLED. Should be quite the unit.
If the Mac Pro is waiting for M3, then likely that "quadro" concept is dead. Pounding round pegs into square holes probably didn't pragmatically work well and that part of the very extended delay is an adjustment to actually using well designed chiplets instead of optimized to fit narrow laptop logic board's packaging.
Something more powerful than a "Ultra class" like package, but slapping together four monolithic laptop chips wasn't a good chiplet design at TSMC N5. It only gets substantively worse at N3 and future fab processes.
Questionable even if the "Ultra" at the N3 stage is going to physically look like the M1/M2 variants of it either.
More POWAAAAAA. I agree 100 percent. throw performance per watt out the window. Really push the power through them to see what they could really do. Maybe the new Mac Pro if it comes will do that. Interesting times!
Something seriously wrong with your desktop pc then, or it's extremely old before PWM fans were a thing. Mine is silent and only emits a very dull low quiet whoosh when working with large video files. Anything else the fans are inaudible. My system is cooled completely by noctua. 19db is the loudest the fans get on max. you can barely hear 19db.
He seems to be stuck on the notion that Apple can't walk away from the 27" (large screen) iMac. He keeps searching for stuff around the Mac Studio to kill off the Studio. The large screen iMac is going to 'kill' it. The Mac Pro is going to 'kill' it.
Apple got on stage over a year ago and effectively explicit said that the large screen iMac was being replaced by the pair of Studios ( Mac + Display). The iMac has been the 'king of the desktops strategy' for Apple for a very long time and he seems to having lots of trouble letting go of that. Pretty good chance Apple already has.
Several reasons.
1. Mac + Display actually lets Apple raise the average selling price.
( yes Apple sells less LCD panels but the margin on the now lower subset they do sell is higher)
The can also mop up the old iMac 27" entry users who are more price senstive with Mini Pro + Display or Mini Pro + 3rd party display. So probably no backslide at all in number of units sold in this old Mac 27" price range at all.
( old iMac Pro mostly covered by Studio Ultra + <some monitor> )
2. If the RAM and GPU are soldered down then end user lifecycle costs are better controlled by detaching the display. For example , Mac Studio get recycled every 3 years and screen every 5-7.
3. Display technology is in flux. Apple switched to the 5K 27" display and then sat on that same tech for almost a decade with no substantive changes. Is Apple's large screen very long term future double layer OLED , miniLED , microLED ... suspect Apple doesn't know. The context to herd Mac users onto the same screen for next 5-6 years just isn't there. Their ability to leverage a screen manufacturing into giving them the discounts for a 2-4 year old screen tech on a bleeding edge screen (because probably going to buy for next 5 years in very , very large numbers) is very limited.
4. Changes in 'Right to repair' and 'mandatory spare parts' regulations. Detaching the monitor is going to help.
25 years ago the iMac bascially saved Apple. It is 25 years later... the iMac is not the primary lynchpin to making this transition successful. Just not. That is likely also a major contributor why the 24" model has been comatose for 3 years. Gurman can't see the forest for the iMac trees.
P.S. there was probably a 27" iMac project rumbling around inside of Apple during the last 3 years or so . Possibly as a backstop in case the Mini/MiniPro/MacStudio didn't work out well. But they have. And most of the troublesome quirks of the Studio Display has largely dropped away also over time.
And 3rd party display alternatives are just getting better. ( At one point Apple made laser printers up until they didn't). I don't think Apple would completely exit , but also don't think Apple is OCD going after the maximum number of desktop deployed screen docking display stations either.
It likely couldn't have been the same exact package as the M1 Ultra if there were 3-6 slots in the Mac Pro. (Unless Apple had resorted to extremely cheesy hackery and was driving those slots with some Rube Goldberg Thunderbolt to internal Thunderbolt back haul.).
The M1 and M2 Max don't have any where near the PCI-e v4 back haul for a Mac Pro. They have four x1 PCI-e v4 lanes. The Ultra config would be a mostly useless, eight x1 PCI-e v4 which would do a whole lot of nothing for x8 and x16 PCI-e v2, v3,v4, or v5 cards.
For their laptop dies Apple pragmatically neutered them on having any kind of respectable PCI-e backhaul provision abilities that the Intel chips had ( even for laptops). Apple declared 'war' on discrete GPUs inside of laptops. Those are all gone now. But that 'war' has collateral damage on being able to provision any kind of reasonable general I/O backhaul for any other usages. ( it is a design trade off , not a 'free lunch' )
Pretty likely it was a incrementally augment package that may have used the laptop dies , but also some augment die to get out of the hole the laptop optimization drags the "cheaper Ultra" into.
Just because the core count is the same doesn't necessarily mean the package contents are the same when there are multiple chip dies present in the package. If can add one die to the package adding two dies isn't a big leap and they all don't have to be monomaniacally exactly the same. ( primarily point of good chiplet design is to have different dies in the package... not dies that all exactly the same. )
If actually trying to practice good chiplet design it doesn't have to be used. If trying to pound a round peg into the square hole ... maybe. However, the problem there is trying to push something into what it really doesn't do well. Better solution there is make something that is better designed in the first place. Square peg for square hole doesn't need as much 'pounding'.
I really don’t understand all the Gurman hate here. Talk about biting the hand that feeds you! MacRumors would be a super boring place without the number one Mac rumor reporter.
He just said on a Bloomberg Twitter space “Yes, I admit I said before that my opinion was that the Studio should get skipped this generation, and I stand by that opinion, but nonetheless the reality is that the Studio is getting updated.” (paraphrased)
He’s said some silly things like that the M3 was coming really soon after the M2. That was obviously wrong, and easy just to ignore. While his opinions like that are sometimes wrong (he’s only human like the rest of us) he undeniably gets the best leaked information of any journalist. When he reports “I got this information and here is the proof” then great, it’s exciting and worth paying attention to. When he says “it seems like this may happen in the future” then you can certainly take that with a grain of salt.
It’s great news that the Studio is getting updated! Are you not excited about that? Is your only reaction really “Gurman is dumb”? What happened to this site. All the hate has made things pretty toxic around here, sheesh.
He just said on a Bloomberg Twitter space “Yes, I admit I said before that my opinion was that the Studio should get skipped this generation, and I stand by that opinion, but nonetheless the reality is that the Studio is getting updated.” (paraphrased)
He’s said some silly things like that the M3 was coming really soon after the M2. That was obviously wrong, and easy just to ignore. While his opinions like that are sometimes wrong (he’s only human like the rest of us) he undeniably gets the best leaked information of any journalist. When he reports “I got this information and here is the proof” then great, it’s exciting and worth paying attention to. When he says “it seems like this may happen in the future” then you can certainly take that with a grain of salt.
It’s great news that the Studio is getting updated! Are you not excited about that? Is your only reaction really “Gurman is dumb”? What happened to this site. All the hate has made things pretty toxic around here, sheesh.
Not necessarily. I have a number of PCs at work we use for CAD. Even with Noctua fans they are noticeably more noisy than the Mac Studios which are in effect silent (below the background noise level in the office). Once we kick in anything that ramps up CPU or GPU use, the fans have to spin up and the PCs are just plain noisy. It's incredibly difficult to ever make a Mac Studio spin up its fans to where you can hear them, and it's never as noisy as any of my PCs.
Laptops...ooof. The Lenovo we use for mobile CAD is so loud it's obnoxious, unless you throttle the performance so hard it's a total dog. Unplug it and you either get terrible performance or the battery blows out in no time at all. When it come to amazing power and very limited energy use and heat dissipation Apple Silicon is just light years ahead of the Wintel world.
I think that was original plan. However, they can't seem to get mac pro ready. Plus any feelers they put out have basically said if it's just a studio in a bigger package, why buy it over a studio, so they need to engineer some kind of upgradability, which they clearly didn't have in mind with the all in one SoC design before. So unless they figure out external memory and/or gpu, mac pro continues to be on hold, so yeah to not show up empty handed they had to revive studio.
I call bull on the noctua comments. Mine are silent unless you have your ear lodged on the case. As for the laptops yep some are obnoxious, mine isn't. But you can hear it, just as you can hear a Macbook Pro M series when stuff starts getting pushed through it.
I think once the update to M2, the whole entry level Mac Studio M2 Max vs M2 Pro Mini will be even more confusing.
These things are not "just" CPU upgrades. The Ultra (which means the Max) required a bunch of non-trivial changes to handle things like coherency, power co-ordination, and interrupt distribution across the two chiplets, along with similar changes to the GPU (to distribute work between the two chiplets).
This functionality was clearly the weakest part of the Ultra, and Apple knew that going in -- but you learn by doing.
There's a long paper trail of patents for a generalized solution to all these issues. It's unclear if they are FULLY implemented in the M2 Max (and then ?M2 Ultra?) My guess is not, and I assumed that Apple would wait till the M3 for a full implementation.
But as I said, you learn by doing! Maybe Apple concluded it was better to ship a 50% improvement solution, ie implement the easiest parts of the full Scalable solution on the M2 framework, and learn from that, rather than the riskier solution of waiting till the M3 and trying to fix everything all at once.
At some point this stuff will be as routine as the iPhone annual chip updates. But right now it is not; every step from Pro to Max to Ultra, then to M2 Ultra, then to M3 ?Extreme? involves massive design effort and validation, and takes time time time. If you consider that Apple are basically rerunning the story of Intel, IBM, and nVidia, at 4x the speed, you'd be a little more impressed and patient!
Here's an example of the multi-chiplet SoC patent: https://patents.google.com/patent/US20230056044A1
This is a LONG patent and as a pending patent the presentation on Google Patents (and everywhere else) kinda sucks :-(
Here's kinda the equivalent for how work is distributed across multiple chiplet GPUs https://patents.google.com/patent/US20230047481A1
Don't worry, there are also some very impressive ray tracing patents. Presumably that's what was dropped from the A16, but the ray patents come in two clusters, so maybe we will get the second cluster with the A17, and even better functionality?
What's confusing? You buy the mini if you want compute, but don't care that much about graphics; you buy the studio if you really care about GPU.
He can opine all day long if he wants. I'm frankly surprised at how nonsensical some of his "opinions" are though. (Really? Apple not wanting to update the Studio because it would cut into Mac Pro sales?) But... I guess anyone can start a rumor. Gurman made a name for himself originally by having some solid contacts / sources for information so people put a lot of stock in what he says. We just have to adjust the amount of salt we take with his comments these days I guess.
I can post photos of the machines...there's no bull here at all. Maybe your PC just isn't ramping up enough to cause the fans to have to ever spin much; mine clearly do. Sure when the PC is loafing along they are quiet but both the Intel and AMD cpus will just use a lot more power and generate a lot more heat than Apple Silicon, and when that happens, the fans do have to spin up and you can hear them. It's physics, there's not an easy way around it. Heck even my "silent" 2019 Mac Pros will get louder when the fans are forced to spin up to cool the AMD GPUs or the Intel Xeon CPU.
I think the issue was people looking at an M2 Pro for Mini. The Pro is effectively as fast as a Max chip on the CPU side, with the Max having more GPU cores. For those of us not just rendering, the M2 Pro is good enough and for many applications where you just rely on CPU speed, the Pro and Max are equivalent. But most higher end spec users aren't going to settle for base RAM or storage either, so once you put in 1TB, 32gb RAM, you have an $1899 Mac Mini versus a $1999 Mac Studio. If you throw in the $100 10 GbE upgrade, then its equivalent, but my point was even at a $100 difference, many want to just jump to Mac Studio. At least you can argue that it's a M1 Max versus M2 Pro, but now if you make Studio M2 based, then that $100 extra is almost a no brainer for most people to skip the Mac Mini and jump to a Mac Studio.
The problem as usual is that once you start upgrading specs and configs, when you have tiered products, you get dangerously close to the next level and that's why there's always talk of cannibalization. The same applies when you start up-speccing an Macbook Air--you start getting close to "might as well buy the base model MacBook Pro."
I don't doubt you own them. I am rebutting the claims the noctua fans are super loud like you say. I am doing multiple 4k and 6k projects at a time. Not light tasks. Some pretty heavy lifting projects. Again, my fans just make a mild woosh at full honk.
Nowhere did I say "super loud." Compared to my Studios though a mild whoosh is loud--they are darned close to dead silent, even under load like working on a 4K project or batch-processing thousands of images.
Unreal Engine 5.2 has native support for Apple Silicon.
It's pretty obvious to me why we don't like him. He pretends to have insider info on Apple products when he doesn't. He makes wild speculations out of thin air and tries to fool people into believing he knows something when he knows no more than we do. And somehow, no matter how many times this happens, people keep saying he has a near perfect track record.
He doesn't add anything positive to the world of Apple rumors when he completely fabricates awful ideas like that the Mac Studio will be discontinued, just because he personally doesn't think it should get an M2... and then falsely passes it off as insider knowledge.
I don't think so. There's a substantial price opening between $1299 (Mini M2 Pro) and $5999 (Mac Pro). Even if we talk about an upgraded Mini to the realm of $1999 or so, there's still a huge window of opportunity. The Studio fills that gap pretty well unless you're suggesting the original plan was then to bring the entry level Pro price way down.
I also think that once you get a design down, refreshing it to handle M2, M3, M4, etc isn't a huge challenge. Look at the low end iPad. Since 2017 or so it got yearly or so refreshes and likely sold millions if not tens of millions of models. The amount of engineering into that probably was far less than when they changed the Pro to use FaceID in 2018 or likely any iPad Pro refresh.
Once they created this mid range price point, it's not hard to update the device to continue sales at this price point.
Only until macos stops running on intel.
once you can no longer build a pc to run macos will have to buy a real mac again. And I don’t do laptops
The era of being able to run pretty much any OS on pretty much any computer sure was great while it lasted